Renewable Energy Focus  Volume 25  June 2018 www.renewableenergyfocus.

com
ORIGINAL RESEARCH ARTICLE

Cycle counting of roller bearing

oscillations – case study of wind turbine
individual pitching system
Matthias Stammler1,*, Andreas Reuter1 and Gerhard Poll2
1
Fraunhofer IWES, Appelstrasse 9A, 30167 Hanover, Germany
2
IMKT, Leibniz Universität Hannover, Welfengarten 1 A, 30167 Hanover, Germany

The amplitude, frequency and number of oscillating movements have a significant impact on the
possible damage modes of rolling bearings. Changes in amplitude change both the fatigue and wear risk
of the raceways. In order to properly calculate the fatigue lifetime of an oscillating bearing, estimate the
risk of surface-induced damage modes and design test programs for such bearings, it is necessary to
understand the characteristics of the oscillating movements. This paper presents a comprehensive
method for the analysis of time series data including position and loads. The method is applied to the
simulated time-series data of the IWT 7.5-164 reference wind turbine. Wind turbine blades typically
experience wide-band dynamic loads due to stochastic wind conditions. Therefore, it is important to use
time series based cycle counting methods.

Introduction oscillatory in nature, as the power output of the blades is at

Beginning with the early works of Palmgren and Lundberg, bear-
ings in rotating applications have been the focus of numerous
research efforts [1]. Oscillating bearings, however, have not been
investigated to a similar depth so far [2]. Oscillating applications
are less common than rotating applications; they include those
occurring unintentionally ones like vibrations of car wheel bear-
ings during ship transport and those executed intentionally like
crane head movements [3]. An increasingly important application
of oscillating bearings are wind turbine blade bearings.
Blade bearings (also called pitch bearings) are large, grease-
lubricated rolling bearings that connect the rotor blades and
the rotor hub of pitch-controlled wind turbines. They allow
turning (pitching) the blades around their primary (z-) axis, see
Figure 1. Turning the blades, changes the amount of lift they
generate. Thus, pitching can be used for both power and load
control of the wind turbine [4].
The largest rotatory movement of a blade is a 90 turn, which is
used to change from power production mode to idling or vice
versa. The 0 and 90 positions in Figure 1 indicate this movement.
Under most circumstances, all movements of pitch bearings are
*Corresponding author:. Stammler, M. (matthias.stammler@iwes.fraunhofer.de)
maximum at the 0 position. As the wind speed and turbulence
are of complex, stochastic nature, this is also the case for the loads
and movements of blade bearings.
Bossanyi presented the benefits of load reduction control mech-
anisms in 2003 [5]. So-called individual pitch controllers (IPC)
control the pitch angle of each blade individually with the aim of
harmonizing the loads of all the blades and reducing the load
peaks of each individual blade. As the load reduction is substantial
and the loss in power output acceptable, most turbine manufac-
turers use or plan to use IPC. Compared with the classical collective
pitch control, IPC leads to different movement patterns of the
blade bearings, with a substantial increase in small oscillations of
0.2–1.5 for an individual amplitude.
The character of the oscillations affects the risk of two possible
damage modes of the bearing rings: rolling contact fatigue (RCF)
and surface-induced damage modes. This is mainly driven by the
formation of a lubricant film between the rolling bodies and the
raceways of the bearing rings. At larger amplitudes and higher
rolling speeds, the formation of a lubricant film that separates
the contact bodies is more likely [6]. A separation of the contact
bodies reduces the risk of surface-induced damage on the raceways.

1755-0084/ã 2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). https://doi.org/10.1016/j.
40 ref.2018.02.004
Renewable Energy Focus  Volume 25  June 2018 ORIGINAL RESEARCH ARTICLE

TABLE 1
Main parameters of the IWT 7.5-164 [14].
Property Symbol Value
Wind class IEC A1
Hub height hhub 119.3 m
Rotor diameter (unconed) Drotor 164 m

ORIGINAL RESEARCH ARTICLE

Cut-in wind speed Vin 3 m/s
Rated wind speed V 11 m/s
Cut-out wind speed Vout 25 m/s
Minimum rotational speed Vmin 5 rpm
Rated rotational speed Vr 10 rpm
Design tip speed ratio lD 8.4
Rotor configuration Upwind, 3-blade
Tower mass mtower 1,467,355 kg
Rotor-Nacelle-Assembly mass mRNA 536,586 kg
Drive train concept Direct drive

oscillation characteristics on damage modes. Existing analytical

FIGURE 1
Blade coordinate system and pitch angle definition.
The most basic approach to integrate oscillations in the calcula-
tion of bearing fatigue lifetime is to sum them and convert the
distance passed into full rotations, which are used for a lifetime
calculations similar to the ISO281 [7,8]. Houpert and later Kotzalas
and Doll presented a method to adapt fatigue lifetime calculations
taking account of the oscillation angles [9,10]. Both these methods
are not applicable for very small oscillation angles, as surface-
induced wear is the dominating damage mode in this case.
Schwack et al. [2] compare different fatigue lifetime calculation
methods [2]. This comparison showed huge differences in the
calculated lifetime for different approaches.
None of the aforementioned publications (except for [2]) men-
tions how to obtain the oscillation cycles from a given data set.
Due to the widespread use of rainflow countings [11,12] for fatigue
calculations, we assume this method to be the standard approach.
As part of the development process, all wind turbines are simu-
lated in aero-elastic simulation tools like Bladed, Fast and HAWC2
according to IEC 61400 [13]. These models integrate the original
pitch controller of the turbine. The movements of the pitch
bearings are included within the time-series outputs of these tools.
While it is also possible to measure the movements on existing
turbines, it is necessary to use simulation data for the development
of new turbine types, new controllers and blade modifications.
Although not the entire turbine lifetime is simulated, the
amount of data created is vast and needs to be organized in order
to be used as input for the subsequent development of individual
components. Typical analytical methods include the abovemen-
tioned rainflow counting, the creation of load bin distributions
and the calculation of damage-equivalent loads for different slopes
of S–N-curves.
In this paper, the simulation data obtained with an aero-elastic
model of the IWT7.5-164 reference turbine [14] is presented in
‘Reference turbine IWT7.5-164 and simulation’ section and used
for the subsequent analysis. The following ‘Effect of movements
on damage modes’ section briefly describes the impact of
methods for time series containing oscillating data are evaluated
for the purpose of blade-bearing movement analysis in ‘Analysis of
oscillations’ section. ‘Program for data analysis’ section describes a
new, comprehensive method for the analysis of blade-bearing
movement data based on the findings of the afore-mentioned
sections. This method is applied to the simulation data of the
IWT7.5-164 reference turbine and the results are presented, both
in ‘Results and discussion’ section. The paper finishes by the
conclusions in ‘Conclusions’ section.
Methods
Reference turbine IWT7.5-164 and simulation
The IWT7.5-164 reference turbine is a nearshore wind turbine with
164 m rotor diameter and a rated power output of 7.5 MW. Its
pitch controller includes IPC-algorithms. The IPC is active at wind
speeds of approximately the rated speed of 11 m/s and above [14].
Table 1 contains the most important properties of the turbine.
The design of the blade bearing has been described by Schwack
et al. [15]. It is a double-row four-point-contact bearing with a
pitch diameter of 4690 mm. The balls have a diameter of 80 mm.
For every degree of rotational movement, the ball covers a rolling
distance x of 20.1 mm on the raceway. The short axis of the
Hertzian contact ellipse (2b, see Figure 4) under typical operating
conditions is in the range of 4–5 mm.
A complete simulation set according to IEC 61400-1 [13] was
executed in the HAWC2 simulation code [16]. The duration of
each simulation was 600 s (except for start/stop and error simula-
tions). 684 fatigue simulations were carried out. The discrete time
steps of the output were 20 ms, resulting in 30,000 values per
channel for each simulation file. The simulation output files in
ASCII format have an overall size of 130 GB. Of the signals avail-
able, the pitch angles of the blades and the loads at the blade roots,
which are the same as the blade-bearing loads, are of particular
interest. Figure 2 shows one exemplary time series. The mean wind
speed of this simulation is 11 m/s, which is the rated wind speed of
the turbine. The movements of one of the blade bearings (signal
‘‘pitch angle’’) and the dominant loads on that bearing (signals
‘‘MyB’’ and ‘‘MxB’’, see also Figure 1) are shown. The pitch angle
movements consist of small oscillations with double amplitude of
up to 4 and of larger movements. The oscillations with larger

41
 ORIGINAL RESEARCH ARTICLE
ORIGINAL RESEARCH ARTICLE
FIGURE 2
IWT7.5-164 wind turbine simulation at 11 m/s mean wind speed.
FIGURE 3
Severe fatigue damage (left) and surface-induced damage (right) on bearing raceways.
amplitudes control the power output of the turbine, whereas the
turbines’ IPC executes oscillations with smaller amplitudes for the
purpose of load reduction.
As the simulated time of 600 s is just a short interval of the
overall lifetime of the turbine, each simulation run of 600 s
represents a longer period. This period is obtained by using the
distribution of the wind speed over 20 years. The IWT7-5164 is
designed for IEC class IA. The IEC 61400-1 [13] contains a descrip-
tion of the site conditions for the different generic classes and the
methods for multiplying the simulation data to obtain the loads
for the full lifetime.
Effect of movements on damage modes
Assuming a bearing is correctly designed and mounted, the two
main damage modes of pitch bearings are rolling contact fatigue
and surface-induced damage modes on the bearing raceways [6].
RCF is the dominant mode when a lubricant film separates the
rolling bodies from the raceway and no metal–metal-contact
occurs. Lundberg and Palmgren first described the fatigue mech-
anisms in rolling bearings [1], today fatigue calculation is
42
Renewable Energy Focus  Volume 25  June 2018
standardized in the ISO 281 [8]. This is the case for long, steady
rotations with relatively high speed, when elastohydrodynamic
conditions apply. The extent of RCF damage is affected, among
others things, by the number of times a rolling body passes a
certain area of the raceway and by the load [17]. The left part of
Figure 3 shows severe RCF damage on the raceway of a four-point
contact bearing.
Oscillations with small amplitudes, however, give rise to surface-
induced damage. Elastohydrodynamic contact conditions are less
likely than mixed lubrication regimes. Surface-induced damage
might appear in the form of False Brinelling and/or Fretting
Corrosion [18]. The right part of Figure 3 shows surface-induced
damage on the raceway of a four-point-contact bearing.
The risk of it occurring, its severity and the speed of damage
progression depend on several parameters. These are, among
others, the amplitude and speed of the oscillations, the load
and lubricant. Whether an oscillation generates surface-induced
damage also depends on the size of the contact patch between the
rolling body and raceway as well as the size of the bearing. Thus,
amplitudes of different bearings should not be compared by means
Renewable Energy Focus  Volume 25  June 2018
of their absolute value in degrees, but rather by the relation
between the rolling distance x and the size of the contact patch
along this direction 2b (see Figure 4).
This relation is also called the amplitude ratio. The findings of
Maruyama and Saitoh for oil-lubricated bearings [19] are used for
the subsequent analysis: In [19], it is shown that an amplitude ratio
of 1.5 is necessary to build up a lubricant film in an oil-lubricated
contact. This ratio is likely to differ with grease-lubricated bear-
ings, and the difference depends on several parameters. Grease is
composed of thickener (soap) and base oil, and the base oil forms
the lubricant film under most conditions. As the oil is contained in
the thickener matrix and the proportion which leaves this matrix
is limited, oil supply to the rolling contact is limited as well and
lubricant film thickness is lower than in purely oil-lubricated
bearings under most conditions. However, thickener is also able
to enter the contact area, and in such a case, it contributes to the
separation of the contact partners. Thus, the above-mentioned
value is taken as an indication for the risk of surface-induced
damage.
FIGURE 4
x/2b dimensionless distance parameter.
FIGURE 5
Analytical methods: rainflow counting, bin counting, damage equivalent loads.
ORIGINAL RESEARCH ARTICLE
Analysis of oscillations
As the size of the output files of the simulation makes it inefficient
to use them as a direct input for component design calculation, the
data have to be processed prior to further design steps. The results
of this processing are used to calculate the fatigue lifetime of an
oscillating bearing, estimate the risk of surface-induced damage
ORIGINAL RESEARCH ARTICLE
modes and design test programs for such bearings.
Common analytical methods include rainflow counting [12],
bin counting of signals and damage equivalent load calculation
[20]. Figure 5 outlines the different approaches of these three
methods, the following figures show detailed views of the indi-
vidual processes. The left part of Figure 5 shows the process of
rainflow counting. The result of a rainflow count is a sorted matrix
containing different classes of amplitudes and the respective
numbers of cycles. The middle part of Figure 5 visualizes the
bin-counting algorithm. This algorithm delivers the total duration
a signal is within a certain range of values. If the number of bearing
rotations is counted as well, the bin counting results in a so-called
Load Revolution Distribution (LRD, not included in Figure 5). The
right part of Figure 6 shows an additional simplification of the
rainflow counting result, the damage equivalent loads (DEL). For a
fixed number of cycles and a given S/N slope, DEL give the
amplitude of oscillations whose damage matches the damage of
the rainflow counting result.
Rainflow counting is widely used to create data input for fatigue
calculations (see the left side of Figure 6). The algorithm treats the
distance between the minimum and the maximum value as a
single flank or half cycle (see number 1 in Figure 6). The name
rainflow count is derived from this interpretation, as the red path
behaves like a water drop flowing down a roof if the figure is turned
through 90 . To avoid double counting of value changes, other
flanks that would interfere with the way of the red path are
intersected by it. This thus creates cycles 2/3, 4/5 and 6/7. The
43
 ORIGINAL RESEARCH ARTICLE
ORIGINAL RESEARCH ARTICLE
FIGURE 6
Rainflow counting (left) vs. range pair counting (right).
FIGURE 7
Damage equivalent load process.
half cycle 1 is relevant for the structural fatigue damage of materi-
als (see [12] for further details). However, with regard to blade-
bearing movements, it is more important to focus on the single
distances x covered by the roller/ball on the raceway. Every
interruption of a movement leads to a breakdown of possible
lubricant films and has to be taken into account. As such, the less
common range-pair counting approach (see the right side of
Figure 6) is more suitable in this case. Rainflow counting over-
estimates the number and amplitude of long cycles (see 1 in
Figure 6) and the number of cycles with relatively small amplitudes
(see 3, 5 and 7 in Figure 6).
In its most basic form, bin counting provides the period of time
that any given signal spends within a set range of values. Combi-
nations of bins, for example, subsequent counting of second signal
values may be used to obtain more information. As no amplitude
information is stored, bin counting as a solitary tool is not valuable
in the analysis of blade-bearing movements. Yet, applying bin
counting as a supplementary method to the range-pair counting is
beneficial, as the combination of the two gives information on the
load level for certain oscillating movements, for example.
The process of DEL creation is also shown in Figure 7. The results
of the rainflow counting are a number of combinations of
44
Renewable Energy Focus  Volume 25  June 2018
amplitudes and cycles (indicated as 1, 2, 3 in the figure). Given
a certain S/N slope, these values are transferred to an arbitrarily
chosen amplitude level that is the same for all values (dotted
arrows in Figure 7). The cycle values can then be summed (S-sym-
bol in Figure 7). The S/N slope is again used to transfer this
summed value to a chosen number of cycles and the correspond-
ing amplitude level. None of these operations changes the
summed fatigue damage to the part. As any information about
single amplitude values is erased in the process, DEL results are not
suitable for the analysis of blade-bearing movements.
Program for data analysis
To take account of the range-pair counting method and additional
bin counting, a MATLAB-program was created for the purpose of
blade-bearing movement and load analysis. Figure 8 shows a flow
chart of this program. The raw time series data is used as the input.
The blade pitch angle signal is analyzed by a range-pair counting
algorithm which includes an evaluation of the mean value of each
half cycle. For each half cycle, additional signals like MxB or MyB
are taken into account via a bin-counting algorithm.
Figure 9 shows and exemplary time series and visualizes the
results of the analysis. The time series file chosen for this example
Renewable Energy Focus  Volume 25  June 2018
FIGURE 8
MATLAB program: simulation time series (left) are processed via range-pair and bin counting into a result matrix.
FIGURE 9
Visualization of the analytical results of an exemplary time series.
was one of the six files (seeds) of DLC1.1 with a wind speed of 9 m/s
and the wind direction of 0 . The pitch angle signal is shown in
blue and contains 13 visually identified half cycles with double
amplitude above 1 . This threshold was set arbitrarily to allow easy
visual identification. The amplitude sizes(in the gray boxes of
Figure 9) are the output of the MATLAB-analysis. The bin counting
result of MxB (red signal) is visually interpreted by means of
colored boxes. The bin counting is executed only when pitch
half-cycles are identified. Time periods without colored boxes
are such periods where no half cycles were identified (either
standstill or too small an amplitude).
Results and discussion
All fatigue load cases according to IEC 61400-1 [13] were analyzed,
namely the Design Load Cases 1.2, 2.4, 3.1, 4.1 and 6.4. The
operating time of these load cases sums up to 175,000 h, that is,
approx. 20 years. Start and stop (DLC 3.1 and 4.1) were assumed to
occur 1100 times each year based on values of comparable wind
turbines.
Table 2 shows the results of a range-pair counting of movements
between 0.03 and 90 width. The lower value is included in the
range, the upper excluded. The overall number of cycles identified
is 566  107. Taking the mean frequency and the numbers of cycle,
it can be calculated that the pitch system is active for 35% of the
ORIGINAL RESEARCH ARTICLE
ORIGINAL RESEARCH ARTICLE
operating time or 60,766 h. This figure corresponds to the IPC
controller mode being active at the rated speed and above, as the
Weibull distribution for this wind class implies a probability of
37% for wind speeds of 10 m/s and above. The high number of
cycles in the smallest amplitude bin (0.03–5 ) is typical for an
individual pitch controlled turbine as well.
As most pitch activity occurs in the bin with the smallest
amplitudes (0.03–5 in Table 2), it is further split up into finer
bins, see Table 3. The amplitude is given both in degrees and
rolling distance covered by the ball on the raceway. Assuming an
amplitude ratio of roughly 1.5 as the lower limit for a lubricant film
formation (cf. ‘Effect of movements on damage modes’ section), a
limit for cycles that impose a risk of surface-induced damage can be
estimated. This limit is only an orientation value, as further
research in the field of oscillating movements of grease-lubricated
contacts is necessary. A value of 0.4 for the double amplitude was
chosen as it was closest to an amplitude ratio of 1.5. The sum of all
cycles with amplitudes equal to or smaller than 0.4 is 1.29  107.
In several research projects dealing with smaller oscillating bear-
ings, surface-induced damage was created with far lower cycle
numbers of up to 0.5  106. A list of research projects and the
kinematic conditions investigated in these can be found in [6].
Thus, the results confirm that there is a risk of surface-induced
damage in the investigated case.
45
 ORIGINAL RESEARCH ARTICLE Renewable Energy Focus  Volume 25  June 2018

TABLE 2
Cycle count results for 0–90 .
Double amplitude ( ) Number of full cycles (–) Mean frequency (Hz)
0.03–5 4.81E + 07 0.43
5–10 8.11E + 06 0.16
10–15 3.31E + 05 0.11
ORIGINAL RESEARCH ARTICLE

15–20 1.97E + 04 0.10

20–25 1.15E + 03 0.05
25–30 4.79E + 03 0.03
30–90 1.76E + 04 0.02

TABLE 3
Detailed cycle count results for 0–5 .
Double amplitude ( ) Double amplitude x (mm) Number of full cycles (–) Mean frequency (Hz)
0.03–0.2 0.6–4 8.66E + 06 0.87
0.2–0.4 4–8 4.25E + 06 0.59
0.4–0.6 8–12.1 3.05E + 06 0.51
0.6–0.8 12.1–16.1 2.32E + 06 0.45
0.8–1.0 16.1–20.1 2.30E + 06 0.40
1.0–1.2 20.1–24.1 2.00E + 06 0.37
1.2 –1.4 24.1–28.2 1.84E + 06 0.33
1.4 –1.6 28.2–32.2 1.63E + 06 0.31
1.6–1.8 32.2–36.2 1.73E + 06 0.30
1.8–2.0 36.2–40.2 1.43E + 06 0.28
2.0–3.0 40.2–60.3 6.37E + 06 0.25
3.0–4.0 60.3–80.5 6.76E + 06 0.21
4.0–5.0 80.5–100.6 5.77E + 06 0.19

TABLE 5
TABLE 4
Bin count of pitch angle mean values. Bin count of resulting bending moment.

Double amplitude ( ) Pitch angle value ( ) Operational time (h) Double Resulting bending Operational
amplitude ( ) moment (kNm) time (h)
0.03–0.40 0–2.5 3450.47
2.5–5 289.81 0.03–0.40 6000–8000 135
5–10 658.67 8000–10,000 205
10–15 553.69 10,000–12,000 286
15–20 198.00 12,000–14000 401
20–25 45.27 14,000–16,000 601
16,000–18,000 928
0.40–1.00 0–2.5 2979.88 18,000–20,000 1116
2.5–5 534.41 20,000–22,000 924
5–10 964.93 22,000–24,000 452
10–15 720.15 24,000–26,000 106
15–20 310.73
20–25 64.01
1.00–5.00 0–2.5 6208.65
2.5–5 3950.09 Table 4 shows the results of this bin count. For double amplitudes
5–10 9480.14 smaller than 0.4 , 66% of the oscillations have a mean value
10–15 9417.40 between 0 and 2.5 . This range contains the largest single operat-
15–20 4595.34
ing time bin in the evaluated values. Sixty-six percent equals
20–25 1290.96
0.86  107 cycles, a number still high enough to impose a risk
of surface-induced damage.
A high number of oscillations with small amplitudes increases The analysis of pitch angle amplitudes indicates the risk of
the risk of surface-induced damage, especially if the mean values surface induced damage for this blade bearing. Whether such
do not change over time. A changing mean value would result in a damage occurs also depends on various other factors, one of them
change of the raceway areas that are in contact with the rolling being the applied load. To gain an understanding of the load
bodies. These areas share the oscillations/loads, thus reducing the conditions, a bin count of the resulting bending moment was
probability that one individual area will fail. To better understand carried out. Table 5 contains the results of this bin counting.
the mean value behavior of the pitch angle, a bin count of the The analytical result is a range pair count including oscillation
pitch angle was performed in addition to the range pair count. frequencies and accompanying bin count of pitch angle and

46
Renewable Energy Focus  Volume 25  June 2018
bending moments. This result can be used as input for a rolling
contact fatigue calculation (see [2]) or to design a test program for
either a test of surface-induced damage modes or rolling contact
fatigue. For fatigue calculations, it is recommended to use a higher
number of load bins, as otherwise the design of the bearing would
be too conservative [21]. The MATLAB-program we use for this
paper can easily extract smaller bin sizes, yet the length of a table
containing several hundreds of bins is too large for the format of
this paper. High bin numbers are not necessarily needed for
estimations on the probability of surface-induced damage modes.
The dependence of such damage modes on outer load has not been
put into calculation methods so far.
Conclusions
Rolling bearings in oscillatory applications may be subject to
different damage modes than bearings in rotating applications.
To better understand the nature of the damage mechanisms of
such bearings, it is necessary to describe the oscillatory move-
ments. In this paper different methods of analyzing movement
data of oscillating bearings were presented. The blade bearings of
the IWT7.5-164 reference wind turbine were used as an example to
demonstrate the methods. The possible damage modes of race-
ways were briefly presented and the amplitude ratio x/2b was used
instead of absolute amplitudes for further analysis. It was shown
that rainflow counting is not a suitable method to analyze move-
ment data, as rainflow counts overestimate the number of both
very small and very large amplitude cycles. A combination of range
pair counting and bin counting was applied to the output data of
the aeroelastic simulations of the reference wind turbine. Taking
account of amplitude ratios that most likely will not allow for
lubricant film formation (x/2b < 1.5), it was found that cycles
with a double amplitude of 0.4 or less bear a risk that surface-
induced damage modes will occur. 0.4 is equivalent to a distance
of approx. 8 mm on the raceways. 1.29  107 cycles with such
amplitudes were identified. In a second step, a bin count of the
pitch angle values showed that 0.86  107 or 66% of the afore-
mentioned cycles are movements around a middle position in the
range of 0–2.5 . If these cycles were in a direct succession, it would
have a duration of approx. 127 days. Furthermore, a bin count of
the resulting bending moments showed that 58% of the bending
moments are within a range of 16–22 MNm. A risk of surface-
induced damage was identified. The results of the analysis can be
used for RCF calculations or to design test programs for different
damage modes.
Future work in this field will aim at a better understanding of the
risk of surface-induced damage for a given bearing system. Part of
ORIGINAL RESEARCH ARTICLE
this work will be a movement pattern analysis of data and incor-
poration of long term wind measurements. Advanced pattern
recognition, in particular, will allow for a better understanding
of critical conditions. The work will also include bearing tests
under different oscillating movements to identify critical ampli-
tude ratios under these conditions.
ORIGINAL RESEARCH ARTICLE
Conflict of interest
None declared.
Acknowledgements
The present work was carried out with the project ‘‘HAPT – Highly
Accelerated Pitch Bearing Tests’’. The project funding by the
German Federal Ministry for Economic Affairs and Energy is kindly
acknowledged.
References
[1] G. Lundberg, A. Palmgren, Dynamic Capacity of Rolling Bearings, Generalstabens
Litografiska Anstalts Förl., Stockholm, 1947.
[2] F. Schwack, M. Stammler, G. Poll, A. Reuter, J. Phys.: Conf. Ser. 753 (2016).
[3] J.O. Almen, J. Mech. Eng. (1937) 415–422.
[4] T. Burton, Wind Energy Handbook, 2nd ed., Wiley, Chichester, New York, 2011.
[5] E.A. Bossanyi, Wind Energy 6 (2) (2003) 119–128.
[6] M. Stammler, G. Poll, Schadensmechanismen in Rotorblattlagern, 2014
Göttingen.
[7] Germanischer Lloyd, Richtlinie für die Zertifizierung von Windenergieanlagen,
Hamburg, 2010.
[8] DIN ISO, Wälzlager – Dynamische Tragzahlen und nominelle Lebensdauer (ISO
281:2007), Beuth, Berlin, 2010.
[9] L. Houpert, Tribol. Trans. 42 (1) (1999) 136–143.
[10] M.N. Kotzalas, G.L. Doll, Philos. Trans. R. Soc. A: Math. Phys. Eng. Sci. 368 (1929)
(2010) 4829–4850.
[11] C.H. McInnes, P.A. Meehan, Int. J. Fatigue 30 (3) (2008) 547–559.
[12] M. Matsuishi, T. Endo, Fatigue of Metals Subjected to Varying Stress, 1968
Fukuoka, Japan.
[13] IEC, Wind Turbines – Part 1: Design Requirements, 3rd ed., International
Electrotechnical Commission, Geneva, 2005.
[14] A. Sevinc, M. Rosemeier, M. Bätge, R. Braun, F. Meng, M. Shan, D. Horte, C.
Balzani, A. Reuter, IWES Wind Turbine IWT-7.5-164, 2014 Hannover.
[15] F. Schwack, M. Stammler, H. Flory, G. Poll, Free Contact Angles in Pitch Bearings
and their Impact on Contact and Stress Conditions,, in: Wind Europe Conference,
Hamburg, 2016.
[16] T.J. Larsen, A.M. Hansen, How 2 HAWC2, The User’s Manual, Risø National
Laboratory, Roskilde, 2015.
[17] E. Ioannides, T.A. Harris, J. Tribol. 107 (3) (1985) 367.
[18] R. Errichello, Tribol. Lubr. Technol. 60 (2004) 34–36.
[19] T. Maruyama, T. Saitoh, Tribol. Int. 43 (8) (2010) 1279–1286.
[20] H.B. Hendriks, B.H. Bulder, Fatigue Equivalent Load Cycle Method: A General
Method to Compare the Fatigue Loading of Different Load Spectrums, 1995.
[21] A.R. Nejad, Z. Gao, T. Moan, Int. J. Fatigue 61 (2014) 116–128.
47